producer_intelligencefandomcom-20200213-history
Alden "Evoke" Mixing Tips Pt.2
Preface Analyzers can be extremely useful in helping to identify problems for some producers, myself included. Ultimately–though–what you're trying to do is train your ears, so rather than using the meters as a crutch, make an effort to use them as a learning tool. Once you have a firm understanding of these concepts, you may find yourself using them less and less until you might not use them at all. Some producers learn to mix without ever using meters. I happen to not be one of those producers. What are "dynamics"? Basically it's the volume of the elements of your tracks, and the ways in which those volumes interact with one another. Why use the word dynamics, rather than just referring to "volume"? The word volume has an oversimplified connotation that isn't particularly helpful when trying to identify and solve dynamics related issues. When people use the word "volume" typically what they mean is Gain. What's the difference? Well "dynamics" contains 3 basic components that are relevant to mixing: Gain, RMS, and Peak. Gain 1.0 Gain is the aspect of dynamics with which most people are the most familiar. This is the easiest to grasp, because it basically just describes how much volume your sound has across the board. More gain = louder sound overall. The peak volume and the average volume (referred to as RMS from here on) increase by the same amount when the gain is increased, and vice versa when the gain is decreased. In this way, gain is very simple. Gain 1.5 Here's the part that a lot of people seem to get mixed up on. Volume is a finite resource. Your track's loudest point is ALWAYS going to be 0. There are no exceptions to this. If you mix with your track clipping and put a limiter on it, it will cap your volume at 0. So from my perspective, it makes the most sense to look at volume as if it's a small box. You're trying to fit all of your sounds into this box, so it doesn't make sense to make one of your sounds the size of the whole box. This is why–to me–the argument for "floating point" volume (where you can technically make anything as loud as you want) is pretty pointless. If you know you're going to have to fit everything in the box eventually, you may as well use the box as a frame of reference while you're assembling the elements of your track. So as you're composing and arranging elements for your song, instead of turning elements up, try to get into the habit of turning elements down. You'll have more space, and your songs will fit in the "box" much more easily. Peak 2.0 Peak volume is also a fairly straightforward concept. When your sound is at its absolute loudest, that's the "peak" volume. So if you have a hat sample, where the initial impact is quite a lot louder than the volume as it rings out, the peak volume is the volume of that first initial "snap" as the drumstick hits the cymbal. Your peak volume is what will register on most volume reading meters. Almost every version of Ableton (save for the most recent) measures peak volume, and not RMS. The same is true across the board for the mixing consoles in most DAWs. Peak 2.5 This next bit is important: MAKE SURE YOU ARE LOOKING AT THE VOLUME OUT–NOT THE VOLUME IN. For a very long time when I was mixing in Ableton, I would read the sliders' relative volumes (for example, reducing the volume of a bass track by 5db) and assume that this meant my bass track was at -5db. This is false. Make sure when you're mixing that you find and read the OUT volume of each track on your mixing console. Samples and synths vary wildly in volume and the amount by which you're adjusting them is meaningless if you don't actually understand the volume you're getting after all your volume processing is finished. Grouping and bussing of tracks is also very helpful here, because it allows you to find the cumulative output volume of all of your synths, for example. RMS 3.0 RMS is an acronym for the way in which it's calculated (Root mean square) but all you really need to know about it is that it's the "average" volume of your sound. Compression increases your sound's overall RMS (without increasing the peak if that's the desired effect) and expansion decreases the RMS (again, sometimes without decreasing the peak, depending). This is something I think is probably best understood intuitively. Usually I'm an advocate for referencing meters to understand the consequences of mixing decisions, but in this case I think RMS has a potential to be more confusing than helpful, and sounds vary so much in their dynamics over time that it'd probably end up making most producers tear their hair out unnecessarily. Nonetheless, understanding that your sound's average volume can be manipulated independently of the peak volume is important. Sometimes you'll read two sounds you're trying to put in the same group, and one will seem louder, but the other will peak at a higher volume. A bit of compression can level out that overenthusiastic peak and bring the two sounds into a bit more dynamic consistency. RMS 3.5 I very rarely read the RMS of a signal, unless I'm limiting my master track. If you want to know how loud your master is relative to most songs, know that -5rms is a LOUD pop mix, and -3rms is a LOUD dance mix. Louder is not better. Too much limiting will distort your sub frequencies, which makes them less subby. Be careful and mindful of your mix. You've spent such a long time making it nice. Don't fuck it up here. I generally shoot for -6RMS to -5RMS, because the amount of side-chaining I do in my tracks generally makes this a non-destructive goal. Understand what effect you're going for and limit accordingly. Your chillout track does not need to be at -4RMS. Closing Notes What we hear as "distortion" typically manifests in the highest frequencies, which is what your ear will usually catch. Distortion in the low end sounds a bit different. The easiest way to explain it is that if you imagine that you've got a sine wave, and you're trying to cap it off to prevent it from being too loud: if you cap it too much, it eventually turns into a square wave. You wouldn't think of a square wave as being a distorted sine, but when it comes to limiting, that's a pretty decent comparison. You want that nice rounded edge to the sine you're using as a sub, because it allows the speaker to bounce back smoothly and push more air in doing so.